ck_pr_fence_load (3) - Linux Manuals

ck_pr_fence_load: enforce partial ordering of load operations

NAME

ck_pr_fence_load - enforce partial ordering of load operations

LIBRARY

Concurrency Kit (libck, -lck)

SYNOPSIS

In ck_pr.h Ft void Fn ck_pr_fence_load void Ft void Fn ck_pr_fence_strict_load void

DESCRIPTION

This function enforces the ordering of any memory load and Fn ck_pr_load 3 operations relative to the invocation of the function. Any store operations that were committed on remote processors and received by the calling processor before the invocation of Fn ck_pr_fence_load is also be made visible only after a call to Fn ck_pr_fence_load . This function always serves as an implicit compiler barrier. On architectures with CK_MD_TSO or CK_MD_PSO specified (total store ordering and partial store ordering respectively), this operation only serves as a compiler barrier and no fence instructions will be emitted. To force the unconditional emission of a load fence, use Fn ck_pr_fence_strict_load . Architectures implementing CK_MD_RMO always emit a load fence.

EXAMPLE

#include <ck_pr.h>

static unsigned int a;
static unsigned int b;

void
function(void)
{
        unsigned int snapshot_a, snapshot_b;

        snapshot_a = ck_pr_load_uint(&a);

        /*
         * Guarantee that the load from "a" completes
         * before the load from "b".
         */
        ck_pr_fence_load();
        snapshot_b = ck_pr_load_uint(&b);

        return;
}

RETURN VALUES

This function has no return value.

SEE ALSO

ck_pr_stall3, ck_pr_fence_atomic3, ck_pr_fence_atomic_store3, ck_pr_fence_atomic_load3, ck_pr_fence_load_atomic3, ck_pr_fence_load_store3, ck_pr_fence_load_depends3, ck_pr_fence_store3, ck_pr_fence_memory3, ck_pr_barrier3, ck_pr_fas3, ck_pr_load3, ck_pr_store3, ck_pr_faa3, ck_pr_inc3, ck_pr_dec3, ck_pr_neg3, ck_pr_not3, ck_pr_add3, ck_pr_sub3, ck_pr_and3, ck_pr_or3, ck_pr_xor3, ck_pr_cas3, ck_pr_btc3, ck_pr_bts3, ck_pr_btr3

Additional information available at http://concurrencykit.org/